Right-leaning think tank says Rove group "misrepresented" its data in ad attacking public employee unions

Original Link: http://voices.washingtonpost.com/plum-line/2011/03/right-leaning_think_tank_says.html

By Greg Sargent

As you may have heard, Crossroads GPS, the group founded by Karl Rove, has gone up on the air with a new 60-second ad that attacks public employees and the Democrats who support them as greedy thugs. The spot, which is backed by a $750,000 buy, represents a national, high-profile effort to turn around a public opinion war the right seems to be losing as the Wisconsin standoff drags on.

But now the fiscally conservative think tank that the ad itself cites as the source for one of its central claims is accusing Crossroads GPS of misusing the think tank's data. The author of the Cato Institute study cited in the ad tells me the spot "misrepresents" his study's findings.

One of the core claims in the ad, which you can watch below, is that unions and Dems are trying "to protect a system that pays unionized government workers 42 percent more than non-union workers." To back up this claim, the ad cites a Cato Institute study from March 2010 that is generally critical of public sector unions. That study is right here.

But the author of the study, Cato director of tax policy studies Chris Edwards, tells me the ad's claim distorts his data in two key ways. The ad says that unionized government workers get paid 42 percent more than non-unionized workers in general, a charge that seems intended to turn non-unionized workers of all kinds against unionized public employees.

In fact, Edwards points out, Cato's study compared unionized government workers only with non-unionized government workers, not with non-union workers overall, and found the first group doing better. In other words, even if the study's overall thrust is critical of public unions, Cato's actual finding on wages would be likely to persuade workers that unions are a good thing -- if you're unionized, you make more than those in the same sector who are not unionized. Instead, the ad misrepresented the finding.

That's not all. Edwards points out that the ad rips the 42 percent figure out of context, further distorting what his study actually found in another way. The study did claim the 42 percent number, but it went on to state specifically that this difference can be partly explained by "general labor market variations across states," because "states with generally higher wages tend to be more unionized."

The study concluded that once you factor in that variable, public sector unions can be said to increase pay levels by approximately 10 percent -- not 42 percent, as the ad claimed.

"The ad misrepresents the gap between union and non-union government workers, and it appears to misrepresent the 42 percent statistic as if it were between government and private workers," Edwards told me.

Crossroads GPS's willingness to use data supplied by a fiscally conservative think tank in a way that even the study's pro-free market author finds objectionable tells you all you need to know about the lengths some conservatives will go to in order to turn public opinion in this fight to their advantage.